How smart is your dog? Can your dog understand images & see what you're pointing at?
It has long been debated by dog owners and scientists at the level of cognitive ability that the domestic dog has. In 2011 some of the questions are answered.
Because many dog owners treat dogs more like humans than dogs, they tend to project human demands and expectations upon dogs. Whether this is in their training methods or just personal bonding, understanding how smart or similar dogs are in their visual processing ability to humans is important. It can dramatically assist owners and trainers in understanding how much dogs can learn.
There have been many experiments made over the last several years, far too many to report in one article, however I will attempt to summarise some of the breakthrough research in this area.
Do dogs understand television or what 2D images mean?
If you are interested in understanding what dogs look at and what they comprehend, a very important first step in this evaluation is how you measure such cognitive abilities. A research project in 2011 at the Helsinki University was developed to understand dog eye tracking in respect to canine cognition. The value of developing an eye tracking device that can evaluate dog’s eye gaze is that it will go a long way towards understanding can dogs process flat images and make mental assessments of what they see.
Essentially their experiment was aimed at understanding if “their gazing behaviour depends on the information on the images.” In this small study of six domestic dogs they were trained to fixate their eyes on an LCD monitor.
“In one test trial six digital colour images (750 x 536 px) of four categories (human faces, dog faces, children’s toys and letters of the alphabets) were displayed for two seconds per frame at grey background. The test trial was repeated three times on one test day, a total of eight test days for five dogs and four days for one dog. Totally 96 different images were shown to dogs. The same picture was repeated 1-5times.In addition, baseline data i.e. dog watching blank screen, was collected during four trials.” Ref 1
The results comprehensively show that dogs CAN process flat images (non-moving) and they pay much more attention to images of other dogs and humans than they do to other images or blank screens.
Specifically the blank screen on average captured the dog’s attention for only 95 ms compared with 259 ms when images were on it. Total fixation time for the four major image types were: images of dogs (534 ms); people (440 ms); toys (300 ms); alphabetic characters (94 ms)
While the above explains the duration of the gaze event and how long it held the dogs interest the images below graphically show the actual results for the four image type categories. On the images the scan paths of two second viewing are recorded with the different colours shown for each test dog.
The circles show a fixation on a particular area with the circle area representing the time hovered in that area. The lines show the eye path taken from one fixation to the next. Very clearly most dogs spent considerable time processing the image of the dogs face and the humans face compared to the human toy and the alphabet image.
How dogs process images of dogs and people
How dogs process images of toys and letters
Dogs recognise human facial expressions
For any dog owner the above title may seem obvious. The more sensitive of dogs out there seem to have a high empathy with us humans and really get what moods we are in. However a dog is a master at body language and change in voice tone, so it may be taking these signals as the strongest indication of our moods rather than reading our faces to know what mood we are in.
Reading of faces has been a human trait since time began and it has been found that woman tend to be much better at this than men. However in the dogs case, reading how other pack animals are going to treat it in the wild, has become a matter of survival for them. So naturally when humans adopt a puppy, the puppy would be very wise to be able to pick up on the signals of its major provider.
The experiment to test the validity of this theory was reported in a journal in January 2011. In the test the researchers showed the dogs photographs of their owners faces on and LCD monitor either with expressions of a smile, anger or being neutral. The researchers were mainly measuring the time each dog looked at each type of facial expression. While the group average did not show a large bias, it was found that dogs generally looked at images of smiles and anger faces for much longer.
They then compared the results with each dogs ‘personality questionnaire’ and the dogs neurotransmitter and hormone genes. They found that here was a strong correlation between the personality type with their genotypes AND the strength of their preference for longer facial study times. This has led to the tentative conclusion that genetic factors in dogs are strong determinants of a dog’s personality type which then affect how they respond to human facial expressions.
This then leads to the concept that particular breeds might be more predisposed when confronted with an owner going through an emotional state to spend more time looking at their owner’s face and trying to figure out what is going on, rather than just listening to their voice or physical behaviours.
Do dogs understand what their owners are pointing at?
This topic has also been hotly contested for many years. Anecdotal evidence has it that many owners would like to think that their dog is very specially aware and can follow simple pointing commands that will make a dog look in a specific direction then follow the hand signal or verbal command such as retrieve. This author can tell you that his dog indeed has very easily picked up the command ‘LOOK’ but unfortunately my dog cannot follow my pointing direction to see a target.
The latest research into ‘comprehension and utilisation of pointing gestures’ was published in May 2011.
In Study 1, the dogs were provided with four potential hiding locations, two on each side of the experimenter to see whether dogs were able to choose the correct location based on a human pointing gesture. In Study 2, the dogs had to rely on a sequence of pointing gestures displayed by two human pointers. This was to test if dogs can understand an ‘indirect signal’. This is where someone motions towards someone else who is pointing at the real object – a very much harder exercise.
The final study was aimed at rating how dogs respond to indirect information about a hidden object (ie pointing at it behind something) and see if the dog can ‘direct the owner to the particular location’.
The results of study one are the most important for most people. It says that dogs can’t “extrapolate precise linear vectors along the pointing arm when relying on human pointing gestures.” Or more simply they don’t understand what you are pointing at in the distance. While people may think that dogs get what pointing means, the researchers believe that dogs visually follow the side of the arm that people are pointing on in a vague directional but alert response and eventually settle upon the target. If there are multiple similar targets on the same side as the owner, the dog will usually show preference for the closest target. More precisely, “dogs may simply go for the side of the human at which they saw a protruding body part and/or movement (Soproni et al. 2002).“ ref 3
The second study suggests that remarkably, dogs can to some extent get an indirect pointing gesture but the this is limited to simple connections. Of further interest is that “dogs are able to localise the hidden object by utilising indirect human signals, and they are able to convey this information to their owner.” Ref 3
It is noted that other studies have found that dogs might be able to follow a simple pointing gesture, but the experimenter’s think that these studies were unintentionally biased. It is believed that the whole human body language and even movement towards a subject has more to do with a dogs success of ‘seeing’ something rather than purely following line of site of where an arm points.
Article by Bruce Dwyer. If you wish to use any of this information please refer to the article as a reference and provide a link to http://www.dogwalkersmelbourne.com.au
If you are interested on being kept up to date on dog article releases or healthy dog treat specials, Join us on http://www.facebook.com/HealthyDogTreatsShop
Ref 1 Eye tracking in canine cognition research – Sanni Somppi. Heini Törnqvist Et al, University of Helsinki 2011
Ref 2 Individual variability in response to human facial expression among dogs Yusuke Hori, Hisayo Kishi, Miho Inoue-Murayama, Kazuo Fujita
Ref 3 Comprehension and utilisation of pointing gestures and gazing in dog–human communication in relatively complex situations. By Gabriella Lakatos · Márta Gácsi · József Topál · Ádám Miklósi